Combination of PAT and mechanistic modeling tools in a fully continuous powder to granule line: Rapid and deep process understanding
András Domokos, Éva Pusztai, Lajos Madarász, Brigitta Nagy, Martin Gyürkés, Attila Farkas, Gergő Fülöp, Tibor Casian, Botond Szilágyi, Zsombor Kristóf Nagy
Abstract
Comprehensive understanding of an integrated continuous pharmaceutical technology was achieved in this study by a combining design of experiments and mechanistic modeling-based simulations. The powder to granule line consisted of twin-screw wet granulation, vibrational fluid-bed drying and milling. A Partial Least Squares (PLS) regression model was built using Near-infrared (NIR) spectroscopy for the real-time monitoring of the product moisture content after the milling step. A split-plot full factorial experimental design was set up and executed to help the understanding of the relationships between the moisture content and process parameters. Furthermore, a mechanistic model was built, involving heat transfer between the drying air and the solid material. The unknown kinetic model parameters were estimated using the results of the experimental study resulting in good calibration and validation performance. The simulations not only reinforced the experimental observations but also paves the way for model-based process monitoring and optimal control.